Automated dealing systems (e.g., for trading currencies, commodities, and the like) are increasingly replacing the conventional manner of dealing using a broker as an intermediary. When a broker is used to complete a transaction, anonymity of the counterparties is preserved either throughout the deal or until just prior to the completion of a transaction depending on the conventions of the particular market. The brokers are familiar with the trading practices of their clients and therefore help to prevent traders who do not want to trade with one another for whatever reason from dealing with one another. Removal of such human safeguards has lead to the development of automated checks and validations in the automated dealing systems.
For example, some known automated trading systems allow traders to enter credit information which is used to check the suitability of counterparties before the deal is completed and before the identity of the parties is revealed. One such system is described in U.S. Pat. No. 5,136,501 wherein, prior to the completion of a transaction, a credit check is performed to insure that each party is willing to extend sufficient credit to its potential counterparty. Another known trading system is described in European Patent Application 92303437.5 in which the system automatically matches offers and bids using credit ranking information entered by each trader.
These and other known trading systems have a number of drawbacks. First, these systems are only amenable to highly specified trading instruments in which all criteria on which a decision to trade is based are readily quantifiable and standardized in the industry and the system. For example, decisions to trade some types of highly specified financial instruments are based solely on the price of the instrument and the quantity available. These easily-defined criteria are easy to incorporate into an automated trading system. However, the known automated trading systems are not capable of accommodating types of financial instruments that are traded using more subjective, less-quantifiable criteria. For example, known automated trading systems do not provide traders with the opportunity to filter out potential deals with other traders who may be unacceptable trading partners on the basis of subjective criteria other than the party's credit, for example, geographic location or political or other competitive criteria. Hitherto, this has only been possible through the agency of a broker who may take into account his client's other types of less quantifiable, subjective criteria concerning parties his clients are willing to deal with while maintaining the anonymity of his clients. Therefore, there is a need for an electronic trading system which accommodates subjective, less-quantifiable trading criteria.
Second, the marketplace may create new, non-standardized types of trading instruments to fit its specific needs. The known electronic trading systems are not capable of accommodating these non-standardized trading instruments because the instrument specifications in these systems are pre-defined based on standardized trading instruments. Therefore, there is a need for an automated trading system which is capable of accommodating non-standardized trading instruments.
Third, in the known automated trading systems, once a trader has entered a bid or offer, the trader no longer has the discretion of negotiating the entered terms of the bid or offer. The system automatically executes trades when compatible offers and/or bids are found. In some systems, a trader may enter a "soft" offer or bid, wherein the trader retains the discretion to either execute or not execute the trade. However, the terms of such a soft offer or bid define the objective criteria that must be satisfied to create a firm offer or bid. The known systems provide no means by which a trader can input a mere "expression of interest" in a particular transaction wherein the trader need not provide predefined objective criteria which would make the expression of interest firm.
In other words, the known trading systems are designed to execute firm transactions when the system locates a bid and offer that match based on detailed specific information concerning the terms of the bid and offer input by the users. These systems do not provide a means by which two parties who are potentially interested in dealing with one another may be introduced to one another based on preliminary information input into the system, and then allowed to negotiate the terms of a transaction using a communication link.
Fourth, the known automated trading systems cannot accommodate credit-complex trading instruments. Credit-complex trading instruments are those for which the calculation of a trading party's risk or exposure at a given time is based on multiple elements and is therefore too complex to integrate into a large-scale trading system. Generally, in order to calculate its exposure, a bank must evaluate several types of risk, for example, credit risk, settlement risk, and liquidity risk. Credit risk is the effect of the transaction on the bank's overall books if the counterparty goes bankrupt before the transaction is completed. Credit risk is evaluated as the replacement value of the transaction assuming that the counterparty is unable to compete the transaction. Settlement risk is the risk that a bank will complete its half of the transaction and the counterparty will be unable to complete its half of the transaction, for example, because the counterparty goes bankrupt prior to settlement. Liquidity risk is the risk that the holder of an instrument will not be able to sell that instrument at a reasonable price when the holder wishes to liquidate the position.
The determination of credit risk is fairly straightforward for short-term transactions such as spot transactions which are settled as soon as the market allows because the risk that a counterparty will go bankrupt during the short period of time prior to settlement is very small. Therefore, it is likely that both parties will complete the settlement of the transaction.
However, the complexity of calculating credit risk increases significantly as the settlement period increases. For example, in forward markets, e.g., the forward foreign exchange and forward rate agreements markets, often transactions do not have a final settlement for several months, a year, or longer. Clearly, there is a greater risk that a counterparty will go bankrupt within this longer period of time prior to settlement. As a result, banks' methods of calculating their long term exposure, including both settlement and credit risk, become increasingly complex and take into account multiple factors.
Therefore, banks and other financial institutions use complex formulae and methods to calculate their potential exposure for each transaction based on a highly complex evaluation of the time decay of the value of money and risk, the institution's total exposure, and numerous other factors. Each financial institution has its own systems and procedures for evaluating its exposure. These credit and risk management procedures are highly confidential and not standardized by any means. As a result, to successfully accommodate these procedures into a single automated trading system, either the financial institutions must standardize their procedures or the implementers of the system must customize their system to accommodate each different institution. Neither of these options is a practicable solution to this problem because banks are not likely to standardize their credit and risk management processes and a customized trading system would be economically infeasible. Also, banks and other trading institutions are extremely protective of information regarding their credit and risk management procedures and may be unwilling to give out this information to third-party programmers who are designing a system or to put this information on line where other parties may be able to access it.